Tree-Ring Stable Carbon Isotope as a Proxy for Hydroclimate Variations in Semi-Arid Regions of North-Central China
Round 1
Reviewer 1 Report
I have read the manuscript by Kang and co-authors with great pleasure and interest. It is a well written manuscript devoted to the climate reconstruction on Central/Northern China based on stable carbon and oxygen isotopes. The narrative is well built around the recent changes on climate in this key region of China. It is rather interesting to see the stronger results obtained with the carbon isotopes when compared to the oxygen isotopes. But the climate signal varies between these two according to several factors, such as species, micro-environmental conditions and overall climate.
There are a few things that I believe that need to be addressed before this manuscript is found acceptable for publication.
What strikes me most is that the R2 values for the verification periods are not so high, but still, the reconstruction underwent by the authors has a strong association with other hydoclimatic reconstructions for this region. Thus, I would recommend the authors to discuss this issue a bit more.
I also noticed another issue in the text. I wonder why the authors present some results at the begging of the discussion section, mainly where you mention figure 4 to 6.
Please, see below some minor comments and suggestions:
Line 54:
I don't fully agree with this statement. They may require detrending to remove ontogenetic trends for instance. Or to account for the Suess effect in the case of the d13C.
Line 80:
Perhaps :"concentrated in the summer (i.e. from June to August)
Line 82:
Do you have a reference to support this observation on the deacresing precipitation trend, or maybe a Figure?
Line 106:
This is a kind of de-trending step. That is why I didn't agree with the statement on de-trending in the Introduction.
Line 120:
The expressed population signal...
Line 120:
It would be also nice to see the average correlation value among the isotope series
Line 155:
Just a suggestion "strengthen when evaluated over three months periods"
Line 181:
The strongest correlation coefficients were observed for the...
Line 187:
between the stable isotopes chronologies and the climate data,
Line 189:
A linear regression model explains...
Line 194:
Maybe I lost this point before, but how did you defined the dorught and wet years? Did you use any criteria for that like certain thresholds?
Figure 4:
This is a nice Figure, but I would re-run the spatial correlation analysis over a slightly larger area. Large enough so that the authors know region covered by the significant correlation values. Alternatively, you can add a panel with a small map highloghting this part of China.
Line 234:
has become more arid with global warming
Line 314:
were measured
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In general, the paper presents properly designed study (however more recent data would be great). The presentation itself requires some improvement.
Majority of "discussion' chapter is actualy the results and should be moved to that part of the manuscript.
The introduction should pay more attention towards to previous similar studies
other minor flaws
fig 2b is wrongly palced, it should be, as Fig. 4, there where scPDSI reconstruction is mentioned (change the numbering of other figs. as a consequence)
in L202 and L326 drought is from 1980 on, while in L224 it is 1990
Author Response
Response to Reviewer 2 Comments
Reviewer #2:
General Comments:
In general, the paper presents properly designed study (however more recent data would be great). The presentation itself requires some improvement.
Point 1: Majority of "discussion' chapter is actualy the results and should be moved to that part of the manuscript.
Response 1: According Reviewer’s suggestion, we combined the two sections for results and for discussion into one section (now as “Results and discussion” in the revised manuscript).
Point 2: The introduction should pay more attention towards to previous similar studies Introduction.
Response 2: We reviewed relevant research literatures published in recent years and added several related sentences and citations in Introduction.
other minor flaws
Point 3: fig 2b is wrongly palced, it should be, as Fig. 4, there where scPDSI reconstruction is mentioned (change the numbering of other figs. as a consequence).
Response 3: Considering the suitable position, we move this Figure to Results and discussion Section in the revised manuscript. To reduce the number of the graphs, we combined the Figure 2a and 2b into one graph in the manuscript. We think that it is appropriate to plot tree-ring chronology and reconstructed scPDSI series together.
Point 4: in L202 and L326 drought is from 1980 on, while in L224 it is 1990.
Response 4: The drought period in L202 and L326 are only present the drought condition of our reconstruction, while in L224 is for the drought of the common period between our reconstruction and other three moisture reconstruction over nearby regions. The moisture reconstruction of Hexi Corridor is start in year 1990. Therefore, the common dry year is 1990 in L224. To make this statement clearer, we added the detailed figure numbers in the relevant sentence in the revised manuscript.
Reviewer 3 Report
This review is about the manuscript entitled “Tree-ring stable carbon isotope as a proxy for hydroclimate variations in semi-arid regions of north-central China” [ID Forests-1597746]. The authors presented a reconstruction of the drought severity index (scPDSI) based upon tree-ring δ13C data. I think that the topic of this study is relevant and very interesting, however, the paper needs to be improved. The weaknesses of this article are mainly the Method and Results sections. My comments on the text are listed below.
- Methods
In general, the methods of analysis should be supplemented and extended
2.2. Sample processing and chronological assessment
- It is necessary to provide information on how the research samples were taken: what were the criteria for selecting trees, how wood samples were collected, at what height of tree the samples were taken? So, I have a question for the authors – whether the samples were taken in accordance with the dendrochronological standard?
- What statistics were used to assess the quality of the Tree-Ring Stable Isotope Chronologies? The authors wrote that the EPS values were calculated, unfortunately, they did not provide any values [lines 120-121- An expressed population signal (EPS) [27] was calculated to determine the strength of the common signal between tree-ring δ13C and the δ18O series]
In my opinion, it is necessary to provide at least the average values of Rbar and EPS for both isotopes.
- What was the sample replication of samples in 1750-1900?
2.3 Climatic data and statistical analyses
lines 134-137 “To identify the presence of wet and dry variations in the climate reconstruction, scPDSI values that were more than one standard deviation (σ) above or below the average value were defined as wet or dry years, respectively”.
I have a question regarding this - What were the reasons for using such a criterion to define dry and wet years? The classification used differs from the scPDSI classification (classification of dry and wet conditions).
See e.g. G. van der Schrier, J. Barichivich, K. R. Briffa, P. D. Jones.2013. A scPDSI-based global data set of dry and wet spells for 1901–2009,
- Results
- In my opinion, Figure 2 should be moved from the Method section to the Results section, The caption to Figure 2 is incorrect, the Figure shows graphs, not maps. Abbreviation RMSE (Figure2b) should be explained.
- In this section, or at least in supplementary Materials, the basic characteristics of the tree-ring δ13C and δ18O chronology should be given.
- There is no information on the Temporal Stability of the Climate Signal. So I have a question, was such an analysis performed? please provide this analysis
- The authors wrote “On an annual timescale, this study period recorded 26 years of drought and 23 years with wet conditions (Figure 2b)”.- Unfortunately, it is hardly visible in the figure, these years should be at least given in additional material
Minor error:
- Lines 167- 169 “ Further comparison of monthly mean scPDSI and SPEI data from May in one year to the following October during 1951-2015 with tree-ring δ13C chronology suggests a stronger correlation than with tree-ring δ18O chronology, except for May to June data for mean SPEI (Figure 3b). - I suppose it should have been added "the previous year."
Best regards,
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Dear Authors,
First, I would like to thank you for the revised manuscript and for taking into account my comments. Generally, the presented work, after corrections and additions, can be considered for publication. I also appreciate the originality of this research.
Slight note to the revised manuscript:
I noticed an editorial error in the revised manuscript - caption for Figure 6 (line 253, that's Figure 46, should be Figure 6).
Best regards,
